A method for predicting oil flow rates is provided. The method includes accessing historical data from a plurality of databases, accessing historical perforation data and historical reservoir properties data from a simulation model, and determining fluid flow values and rock quality index values associated with perforated intervals of the plurality of wells. The method further includes corresponding the fluid flow values and rock quality values to the well production data, training, using the plurality of input values, a machine learning model for predicting oil flow values at perforated intervals of a plurality of target wells, predicting, using the trained machine learning model, the oil flow values at the perforated intervals of the plurality of target wells, and generating a synthetic production log that includes the predicted oil flow values at the perforated intervals of the plurality of target wells.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The method of claim 1, wherein the simulation model corresponds to a reservoir simulation model.
Reservoir simulation is employed to model the behavior of subsurface geological formations, often for oil and gas extraction. A challenge in reservoir simulation is accurately representing complex geological structures and fluid flow dynamics. This invention addresses this challenge by employing a simulation model that specifically corresponds to a reservoir simulation model. This type of model is designed to capture the intricate interactions of fluids (such as oil, gas, and water) within porous and permeable rock layers. The reservoir simulation model can incorporate parameters like porosity, permeability, fluid properties, and well locations to predict how the reservoir will perform over time, including production rates and pressure changes. This specialized simulation approach allows for a more refined and accurate understanding of subsurface reservoir behavior.
3. The method of claim 1, wherein the historical reservoir properties data are associated with one or more of porosity values, permeability values, well geometry, rock classifications, and stratigraphic zone values.
This invention relates to reservoir simulation and modeling, specifically improving the accuracy of reservoir property predictions by incorporating historical reservoir data. The problem addressed is the lack of precise reservoir property data, which leads to unreliable simulations and suboptimal decision-making in oil and gas extraction. The solution involves a method that enhances reservoir modeling by integrating historical reservoir properties data, such as porosity values, permeability values, well geometry, rock classifications, and stratigraphic zone values. These properties are used to refine reservoir models, ensuring more accurate predictions of fluid flow, pressure distribution, and reservoir behavior. The method leverages this historical data to adjust and validate reservoir simulations, reducing uncertainties in reservoir characterization. By incorporating these detailed property values, the system improves the reliability of reservoir performance forecasts, aiding in better resource management and extraction strategies. The approach ensures that reservoir models are grounded in real-world data, leading to more informed and efficient decision-making in hydrocarbon recovery processes.
4. The method of claim 1, wherein the perforated intervals of the plurality of wells and an additional plurality of wells are associated with a plurality of depth values such that each perforated interval is associated with a depth value of the plurality of depth values.
This invention relates to oil and gas well drilling and production, specifically to managing perforated intervals in multiple wells to optimize resource extraction. The problem addressed is the need to accurately track and correlate perforated intervals across multiple wells to improve production efficiency and reservoir management. The invention provides a method where perforated intervals in a primary set of wells and an additional set of wells are each assigned specific depth values. These depth values allow for precise mapping and comparison of perforated intervals across different wells, enabling better coordination of production activities. The method ensures that each perforated interval is uniquely identified by its depth value, facilitating data analysis and decision-making for reservoir management. This approach helps in optimizing well performance, reducing operational inefficiencies, and enhancing overall production output. The depth-based association of perforated intervals allows for improved monitoring and adjustment of production strategies, ensuring that resources are extracted more effectively. The invention is particularly useful in complex reservoir environments where multiple wells are involved, and accurate depth correlation is critical for successful production management.
5. The method of claim 1, wherein the machine learning model is trained on one or more of a GBM algorithm, a random forest algorithm, a tree ensemble algorithm, and XGBoost algorithm.
This invention relates to machine learning systems for predictive modeling, specifically addressing the challenge of improving model accuracy and robustness by leveraging diverse ensemble-based algorithms. The method involves training a machine learning model using one or more ensemble techniques, including Gradient Boosting Machines (GBM), Random Forest, Tree Ensemble, and XGBoost. These algorithms are designed to enhance predictive performance by combining multiple decision trees or weak learners, reducing overfitting and improving generalization. GBM iteratively builds models to correct errors from previous iterations, while Random Forest and Tree Ensemble methods aggregate predictions from multiple trees trained on different subsets of data. XGBoost, an optimized gradient boosting framework, further refines performance through regularization and parallel processing. The approach allows for flexible model training, enabling selection of the most suitable algorithm or combination of algorithms based on the specific requirements of the application, such as computational efficiency, interpretability, or accuracy. This method is particularly useful in applications requiring high predictive accuracy, such as fraud detection, risk assessment, and recommendation systems.
7. The method of claim 6, further comprising determining contribution fraction values for additional perforated intervals of each of the additional plurality of wells.
This invention relates to oil and gas well production optimization, specifically improving hydrocarbon recovery by analyzing and adjusting production contributions from multiple perforated intervals in a well. The problem addressed is the difficulty in accurately determining how much each perforated interval in a well contributes to overall production, which is critical for optimizing well performance and maximizing hydrocarbon recovery. The invention provides a method to calculate contribution fraction values for perforated intervals in a well, which involves measuring production rates and pressures at different intervals and using these measurements to determine the relative contribution of each interval to the total production. The method further extends this analysis to multiple wells, allowing for a comprehensive assessment of production contributions across a field. By determining these contribution fractions, operators can make informed decisions about well management, such as adjusting production rates or targeting specific intervals for stimulation or intervention. This approach helps optimize production efficiency, reduce operational costs, and enhance overall field recovery. The invention is particularly useful in complex reservoirs where multiple intervals contribute to production, and precise control over each interval is necessary to maximize output.
9. The method of claim 8, further comprising dividing the interval flow value of each of the additional perforated intervals of each of the additional plurality of wells by the total flow value of each well that corresponds to each perforated interval.
This invention relates to oil and gas well production optimization, specifically improving flow analysis in multi-well systems. The problem addressed is accurately determining the contribution of individual perforated intervals within each well to the total production flow, which is critical for optimizing well performance and resource allocation. The method involves analyzing flow data from multiple wells, each with multiple perforated intervals. For each well, the total flow value is calculated by summing the flow values of all its perforated intervals. Then, for each perforated interval in each well, the interval flow value is divided by the total flow value of that well to determine the relative contribution of that interval. This process is repeated for all wells in the system, providing a normalized measure of each interval's production efficiency. The method also includes handling additional wells and perforated intervals by applying the same calculation steps to ensure consistent analysis across the entire well system. By quantifying each interval's contribution, operators can identify high-performing or underperforming zones, enabling targeted interventions to maximize overall production efficiency. The approach is particularly useful in complex reservoir systems where multiple wells interact and where precise flow distribution data is essential for decision-making.
11. The non-transitory computer-readable medium of claim 10, wherein the simulation model corresponds to a reservoir simulation model.
A system and method for optimizing reservoir simulation models using machine learning techniques. The technology addresses the challenge of efficiently simulating fluid flow and reservoir behavior in subsurface formations, which is computationally intensive and time-consuming. Traditional reservoir simulation models require significant computational resources and expertise to develop and run, limiting their practical application. The invention involves a machine learning-based approach to accelerate and optimize reservoir simulations. A simulation model, specifically a reservoir simulation model, is trained using machine learning algorithms to predict reservoir behavior under various conditions. The system includes a training module that processes historical reservoir data, such as pressure, temperature, and fluid properties, to train the machine learning model. The trained model can then generate predictions or simulations of reservoir performance, reducing the need for full-scale numerical simulations. The reservoir simulation model is integrated with a machine learning framework to improve accuracy and efficiency. The system may also include a validation module to assess the performance of the trained model against known reservoir data, ensuring reliability. By leveraging machine learning, the invention enables faster and more cost-effective reservoir simulations, supporting better decision-making in oil and gas exploration and production.
12. The non-transitory computer-readable medium of claim 10, wherein the historical reservoir properties data are associated with one or more of porosity values, permeability values, well geometry, rock classifications, and stratigraphic zone values.
This invention relates to a computer-implemented system for analyzing and managing reservoir properties in subsurface geological formations, particularly in the oil and gas industry. The system addresses the challenge of efficiently storing, retrieving, and analyzing large datasets related to reservoir characteristics to optimize exploration and production operations. The invention involves a non-transitory computer-readable medium storing instructions that, when executed, enable a computing device to process historical reservoir properties data. These data include porosity values, permeability values, well geometry, rock classifications, and stratigraphic zone values. The system organizes and correlates these parameters to enhance decision-making in reservoir modeling, simulation, and field development planning. By integrating these diverse datasets, the system provides a comprehensive understanding of subsurface conditions, improving accuracy in predicting reservoir behavior and resource estimation. The technology leverages computational methods to handle complex geological data, ensuring that reservoir engineers and geoscientists can access and interpret critical information efficiently. This approach supports better reservoir management, reducing uncertainties in production forecasts and optimizing drilling strategies. The system's ability to associate multiple reservoir properties allows for more precise reservoir characterization, which is essential for maximizing hydrocarbon recovery and minimizing operational risks.
13. The non-transitory computer-readable medium of claim 10, wherein the perforated intervals are associated with a plurality of depth values such that each perforated interval is associated with a depth value of the plurality of depth values.
This invention relates to a computer-implemented system for managing perforated intervals in a subsurface formation during drilling or well completion operations. The system addresses the challenge of accurately tracking and controlling perforations at specific depths to optimize fluid flow and reservoir access. The invention involves a non-transitory computer-readable medium storing instructions that, when executed, cause a processor to associate each perforated interval with a unique depth value from a predefined set of depth values. This ensures precise correlation between perforations and subsurface locations, enabling targeted reservoir stimulation or production. The system may also include instructions for generating a perforation plan, adjusting perforation parameters based on real-time data, and validating the accuracy of depth associations. By linking perforations to specific depth values, the invention improves operational efficiency, reduces errors, and enhances reservoir management. The system may integrate with drilling or completion tools to automate perforation operations while maintaining depth accuracy. This approach is particularly useful in complex wellbore environments where precise depth control is critical for maximizing hydrocarbon recovery or injection efficiency.
14. The non-transitory computer-readable medium of claim 10, wherein the machine learning model is trained on one or more of a GBM algorithm, a random forest algorithm, a tree ensemble algorithm, and XGBoost algorithm.
This invention relates to machine learning systems for predictive modeling, specifically addressing the challenge of selecting and training effective machine learning models for improved accuracy and performance. The system involves a non-transitory computer-readable medium storing instructions that, when executed, configure a computing device to train a machine learning model using one or more ensemble-based algorithms. These algorithms include Gradient Boosting Machines (GBM), Random Forest, Tree Ensemble, and XGBoost, which are known for their ability to handle complex datasets and reduce overfitting. The trained model is then deployed to make predictions on new data, leveraging the strengths of ensemble methods to enhance predictive accuracy. The system may also include preprocessing steps to prepare input data, such as feature scaling or normalization, and post-processing steps to refine predictions. The use of multiple ensemble algorithms allows the system to adapt to different types of data and problem domains, ensuring robust performance across various applications. This approach is particularly useful in fields requiring high-precision predictions, such as finance, healthcare, and industrial automation.
16. The non-transitory computer-readable medium of claim 15, wherein the stored instructions, when executed by the one or more processors of the computing device, further cause the computing device to determine contribution fraction values for additional perforated intervals of each of the additional plurality of wells.
This invention relates to a system for optimizing hydrocarbon production from multiple wells in a reservoir, particularly focusing on perforated intervals where production occurs. The problem addressed is the need to accurately model and allocate production contributions from different perforated intervals in a well, which is critical for reservoir management and enhanced oil recovery. The system uses a computing device with stored instructions to process well data, including pressure and flow measurements, to determine how much each perforated interval contributes to overall production. The system further analyzes additional wells in the reservoir, calculating contribution fraction values for perforated intervals in those wells. These values help operators understand production distribution across the reservoir, enabling better decision-making for well management and reservoir optimization. The system may also incorporate additional data, such as geological or fluid properties, to refine the contribution calculations. The goal is to provide a data-driven approach to improve production efficiency and resource recovery.
18. The non-transitory computer-readable medium of claim 17, wherein the stored instructions, when executed by the one or more processors of the computing device, further cause the computing device to divide the interval flow value of each of the additional perforated intervals of each of the additional plurality of wells by the total flow value of each well that corresponds to each perforated interval.
This invention relates to data processing in the oil and gas industry, specifically for analyzing well performance. The problem addressed is the need to accurately assess the productivity of individual perforated intervals within multiple wells by normalizing flow data. The system processes flow data from a plurality of wells, each with multiple perforated intervals. For each well, the system calculates a total flow value by summing the flow values of all perforated intervals. Then, for each perforated interval, the system divides the interval flow value by the total flow value of its corresponding well. This normalization allows for comparative analysis of interval productivity across different wells, accounting for variations in overall well performance. The method involves storing flow data, performing mathematical operations to derive normalized values, and outputting the results for further analysis. The invention improves decision-making in reservoir management by providing a standardized way to evaluate the contribution of each perforated interval to the overall well output. This approach helps identify high-performing and underperforming intervals, optimizing production strategies. The system is implemented using a computing device with processors and non-transitory computer-readable storage, ensuring efficient and scalable data processing.
20. The method of claim 19, further comprising generating a synthetic production log that includes the oil flow values at the perforated intervals of the plurality of target wells.
This invention relates to oil and gas well production monitoring and analysis. The technology addresses the challenge of accurately assessing oil flow rates at specific perforated intervals in multiple target wells, which is critical for optimizing production and reservoir management. The method involves analyzing production data from the wells to determine oil flow values at these perforated intervals. Additionally, the method generates a synthetic production log that consolidates these oil flow values for the target wells. This synthetic log provides a comprehensive overview of production performance across the wells, enabling better decision-making for reservoir management and production optimization. The method may also include steps such as processing well data, identifying perforated intervals, and calculating flow rates based on the production data. The synthetic production log serves as a valuable tool for engineers and operators to assess well performance and make informed adjustments to enhance production efficiency.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 28, 2021
April 9, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.